End of training

README.md

@@ -1,9 +1,15 @@
 ---
 base_model: ondevicellm/tinyllama_mole_sft_ultrachat_ep3
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
+- trl
+- dpo
+- generated_from_trainer
+datasets:
+- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: tinyllama_mole_dpo_ep3
   results: []
@@ -14,7 +20,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # tinyllama_mole_dpo_ep3
 
-This model is a fine-tuned version of [ondevicellm/tinyllama_mole_sft_ultrachat_ep3](https://huggingface.co/ondevicellm/tinyllama_mole_sft_ultrachat_ep3) on the None dataset.
+This model is a fine-tuned version of [ondevicellm/tinyllama_mole_sft_ultrachat_ep3](https://huggingface.co/ondevicellm/tinyllama_mole_sft_ultrachat_ep3) on the HuggingFaceH4/ultrafeedback_binarized dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.6285
 - Rewards/chosen: -0.3050

config.json

@@ -36,6 +36,6 @@
   "torch_dtype": "bfloat16",
   "transformers_version": "4.37.0",
   "trust_remote_code": true,
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32000
 }